1 Dynamically Disordered Lattice in a Layered Pb-I ... - ACS Publications

geometry. Page 2 of 33. ACS Paragon Plus Environment. Journal of the ... The examples given above suggest dynamically flexible lattices in 2D and ... ...
1 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OF DURHAM

Article

Dynamically Disordered Lattice in a Layered Pb-I-SCN Perovskite Thin Film Probed by 2D IR Spectroscopy Jun Nishida, John P Breen, Kurt P Lindquist, Daiki Umeyama, Hemamala I. Karunadasa, and Michael D. Fayer J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b03787 • Publication Date (Web): 19 Jul 2018 Downloaded from http://pubs.acs.org on July 19, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Dynamically Disordered Lattice in a Layered Pb-I-SCN Perovskite Thin Film Probed by 2D IR Spectroscopy Jun Nishidaa), John P. Breen, Kurt P. Lindquist, Daiki Umeyamab), Hemamala I. Karunadasa, and Michael D. Fayerc) a)

Department of Chemistry, Stanford University, Stanford, CA, 94305 present address: Department of Physics, Department of Chemistry, and JILA, University of Colorado Boulder, Boulder, CO, 80309 b) present address: National Institute for Materials Science (NIMS) 1-2-1 Sengen, Tsukuba, Ibaraki, 305-0047, Japan c) corresponding author, email: [email protected] phone: 650 723-4446 Abstract

The dynamically flexible lattices in lead-halide perovskites may play important roles in extending carrier recombination lifetime in 3D perovskite solar-cell absorbers and in exciton self-trapping in 2D perovskite white-light phosphors. Two-dimensional infrared (2D IR) spectroscopy was applied to study a recently reported Pb-I-SCN layered perovskite. The Pb-ISCN perovskite was spin-coated on a SiO2 surface as a thin film, with a thickness of ~100 nm, where the S12CN‒ anions were isotopically diluted with the ratio of S12CN:S13CN=5:95 or 1:99 to avoid vibrational coupling and excitation transfer between adjacent SCN‒ anions. The minor S12CN‒ component was the principal vibrational probe that reported on the structural evolution through 2D IR spectroscopy. Spectral diffusion was observed with a time constant of 4.1 ± 0.3 ps. Spectral diffusion arises from small structural changes that result in sampling of frequencies within the distribution of frequencies that comprise the inhomogeneously broadened absorption spectrum of the in the Pb-I-SCN layered perovskite. These transitions among discrete local structures are distinct from oscillatory phonon motions of the lattice. To accurately evaluate the structural dynamics through measurement of spectral diffusion, the vibrational coupling between adjacent SCN‒ anions had to be carefully treated. Although the inorganic layers of typical 2D perovskites are structurally isolated from each other, the 2D IR data demonstrated that the layers 1  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of the Pb-I-SCN perovskite are vibrationally coupled. When both S12CN‒ and S13CN‒ were pumped simultaneously, cross-peaks between S12CN and S13CN vibrations and an oscillating 2D band shape of the S12CN‒ vibration were observed. Both observables demonstrate vibrational coupling between the closest SCN‒ anions, which reside in different inorganic layers. The thin films and the isotopic dilution produced exceedingly small vibrational echo signal fields; measurements were made possible using the near-Brewster’s angle reflection pump-probe geometry.  

2  

ACS Paragon Plus Environment

Page 2 of 33

Page 3 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

I. Introduction Three-dimensional (3D) lead halide hybrid perovskites have attracted a great deal of recent interest for their applications in solar cells.1-3 The low cost for processing perovskitebased solar cells is particularly attractive, rendering them promising candidates for replacing silicon-based solar cells. While the stability and performance lifetime of large-area cells must be improved for practical applications,4-5 the efficiencies of the perovskite-based solar cells have already surpassed 20%.6-7 Furthermore, the two-dimensional (2D) analogs within the lead-halide perovskite family have been studied as green/blue phosphors since the 1990s,8 with more recent work showing white-light emission.9 Several studies have probed the ground-state and electronic excited-state lattice dynamics in halide perovskites.10-13 The outstanding photovoltaic performance of 3D lead halide perovskites has been attributed, at least in part, to their dynamical flexibility. In 3D Pb-I perovskites, optical excitation generates free electrons and holes as carriers.14-18 Some studies have suggested that the local lattice surrounding these carriers reconfigures to stabilize the carriers and impede recombination, leading to the unusually long carrier diffusion length in these absorbers despite their direct bandgap. The result is high power conversion yields.12-13, 19-20 Layered or 2D perovskites have a much higher exciton (excited electron-hole pair) binding energy than their 3D congeners, owing to the spatial confinement of the inorganic layers and the low dielectric constant of the organic layers.21 More pronounced excited-state lattice distortions have been proposed as the origin of white-light emission upon UV excitation from 2D Pb-Br and Pb-Cl perovskites.9 In these 2D perovskites, optical excitation generates excitons that are thought to self-trap through transient lattice distortions, leading to a very broad and Stokes-shifted emission.22 Although self-trapping can occur in static or flexible lattices, dynamic lattice

3  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

fluctuations may allow the exciton to sample a more extensive landscape of energy minima and promote exciton localization and stabilization through lattice distortions. The examples given above suggest dynamically flexible lattices in 2D and 3D halide perovskites. Recently there have been a few reports that directly observe the thermal fluctuations of the lattices in single crystals of 3D perovskites at room temperature. Yaffe et al. described broad low-frequency Raman spectra centered at zero-frequency in single crystals of (CH3NH3)PbBr3 and CsPbBr3,11 indicating that the lattice constantly undergoes thermal fluctuations. Miyata et al. observed the structural relaxation of the same 3D perovskites in the time-domain using time-resolved optical Kerr effect (TR-OKE) spectroscopy, and the relaxations were interpreted in the context of “large polaron” formation.23 The ability of the perovskite lattices to reconfigure on a fast time scale is believed to be important for stabilizing photogenerated carriers.19 Here, we show that a thin film of a 2D perovskite, (CH3NH3)2[PbI2(SCN)2] (Figure 1A), exhibits fast structural changes using two-dimensional infrared (2D IR) spectroscopy. 2D IR spectroscopy has been successful in characterizing the structural dynamics of a variety of flexible materials.24-30 While 2D NMR spectroscopy (such as EXSY) characterizes dynamics occurring in microseconds or slower time scales, 2D IR spectroscopy can directly observe the structural dynamics occurring in sub-picosecond to nanosecond time scales. The long time limit is determined vibrational lifetime of the mode used as the vibrational probe. In the experiments presented here, the lifetime limits the measurements to a few tens of picoseconds. 2D IR spectroscopy monitors the temporal evolution of a molecular vibrational frequency within an inhomogeneously broadened vibrational spectrum. As the vibrational frequency is sensitive to the environment interacting with a molecule, fluctuations in structure lead to changes in the

4  

ACS Paragon Plus Environment

Page 4 of 33

Page 5 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

probe’s vibrational frequency.31-33 Therefore, a measurement of the time dependence of the vibrational frequency reports on the time evolution of the different environments that give rise to the inhomogeneously broadened absorption spectrum of the vibration under study. Figure 1A shows the crystal structure of the (CH3NH3)2[PbI2(SCN)2] layered perovskite.34-36 This perovskite is referred to as the Pb-I-SCN perovskite hereafter. The Pb-ISCN perovskite has a distinct two-dimensional layered structure, with corner-sharing octahedra containing central Pb2+ ions coordinated to bridging iodides and terminal thiocyanates. Small CH3NH3+ cations reside close to the anionic inorganic layers, providing charge balance. The lattice is highly compressible with a particularly low ambient-pressure bulk modulus (K0) of 7.6(8) GPa.35 In contrast, the K0 values of oxide perovskites are greater than 100 GPa.37 Given the soft nature of the material, we postulated that the lattice may undergo fast structural fluctuations. Even though the organic cations are fixed in the lattice, our measurements reveal that the lattice structure undergoes diffusive thermal fluctuations, which were observed as a decay in the time correlation of the CN vibrational frequency (spectral diffusion). Such ultrafast dynamical flexibility found in the 2D perovskite structure can be important for the immediate stabilization of the photo-generated excitons. The diffusive fluctuations of local structures are distinct from phonons, the oscillatory motions of structure about the average structural configuration. The performance of the perovskite-based devices is well-known to be strongly dependent on the morphology of the crystals within the devices.38 It is thus useful to study perovskite structural dynamics in the morphology relevant to the devices, i.e., thin films. A challenge of applying 2D IR spectroscopy to perovskite films with the thickness of ~100 nm is the weak signal produced by the small number of vibrational chromophores in the thin films. In the study

5  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

presented here, this problem is compounded by isotopic dilution of the CN chromophores, which is necessary to extract the dynamical information of interest. Using standard methods, the observed signal level is orders of magnitude smaller than bulk solution samples commonly studied with 2D IR spectroscopy. The employment of BOXCARS geometry with an external local oscillator32 alleviates this problem to a significant extent, but the slow data acquisition rate and phase ambiguity hinder the detailed observations of the structural fluctuations. A new approach, 2D IR spectroscopy with near-Brewster’s angle reflection pump-probe geometry, has proven useful for the study of molecular monolayers.39 This geometry essentially combines (1) phase stability achieved by acousto-optic modulator (AOM) pulse shaping system and self-heterodyne between a local oscillator (LO) and a vibrational echo signal,40-41 and (2) reduction of the LO amplitude relative to the vibrational echo signal amplitude resulting in high signal-to-background ratio. These two important features yield high quality two-dimensional band shapes with fast data acquisition rates even with samples that are thin films or even monolayer.39, 42 2D IR spectroscopy requires vibrational probes in the system of interest, and the CN stretching mode of the SCN‒ anions was used as the vibrational probe.43 As seen in Figure 1A, an SCN‒ has another SCN‒ from an adjacent sheet in close proximity, which, as shown below, results in vibrational couplings and vibrational excitation transfer. These two factors can complicate the interpretation of the observed two-dimensional band shapes. To address these issues, we isotopically diluted the SCN anions44 (S12CN:S13CN=5:95) and controlled the spectral profile of the pump pulses (pulses 1 and 2 in the three pulse 2D IR pulse sequence).45

6  

ACS Paragon Plus Environment

Page 6 of 33

Page 7 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Figure 1. A. The structure of the 2D perovskite (CH3NH3)2PbI2(SCN)2 (abbreviated as Pb-I-SCN). Turquoise – Pb, Purple – I, Yellow – S, Gray – C, Blue – N, H atoms omitted for clarity. The CN stretching modes of two adjacent SCN anions (indicated by dotted red lines) are coupled by transition dipole-transition dipole interactions. B. The sample configuration drawn together with incident and output laser pulses for reflection enhanced 2D IR spectroscopy. The sample film (red layer) was spin-coated onto a 100-nm SiO2 layer, which was deposited on a 3-mm CaF2 window. The sample was in a N2-filled glove box, and an O-ring and another CaF2 window were placed on the top to isolate the sample film from atmospheric moisture. Two pump pulses and one probe pulse were sent through the sample film, where the incident angle of the probe pulse, 53°, was set close to the Brewster’s angle (54.5°). The small probe reflection as the local oscillator, together with the signal field, was sent to a spectrograph and detected by a 32-element HgCdTe array detector.

This paper is organized in the following manner. In Section II, the sample preparation and the configuration of the laser system are summarized. In Section III, we discuss the infrared absorption spectra of the prepared perovskite thin films, the pump-probe signal, and two dimensional band shapes. A significant enhancement in the signal-to-background ratio in the reflection geometry and a contribution from dispersive line shape were both confirmed for the thin film sample, as predicted in a previous publication.39 In Section IV, the 2D IR spectra acquired with the full pump spectral profile are discussed. The observed cross peaks between 12

CN and 13CN stretching modes and the characteristic oscillation of the 2D band shape of the

12

CN mode confirm the vibrational coupling between the adjacent anions. The transition dipole-

transition dipole coupling mechanism was demonstrated by a calculation based on the transition dipole amplitude and the distance and angle between the anions obtained from the crystal structure.34 Then in Section V, the evolution of the 2D band shape observed by pumping the 7  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

12

CN mode but not the 13CN mode is discussed. The band shape observed in this manner is free

from influences of vibrational coupling or excitation transfer and reflects the structural evolution of the system. II. Experimental Procedures A. Isotopically Diluted Pb-I-SCN Perovskite Thin Film Preparation The Pb-I-SCN perovskite was spin-coated on a 100 nm SiO2 film deposited on a 3 mm thick CaF2 substrate (1” diameter) by plasma-enhanced chemical vapor deposition (UCSB Nanofabrication Facility).46 The SiO2/CaF2 wafer transmits incident infrared pulses and was useful for the optical alignment as briefly discussed below in Section IIB. The Pb-I-SCN perovskite was prepared based on a previously published procedure35 with some modifications. Potassium thiocyanate (KS12CN and KS13CN), lead(II) nitrate (Pb(NO3)2), methylammonium iodide ([CH3NH3]I), and tetrahydrofuran (THF) were purchased from SigmaAldrich and used without further purification. In the entire processes below, the exposure to light was minimized by either wrapping the solution/solid containers with aluminum foil or turning off lights. 4.1 mg of KS12CN and 86.5 mg of KS13CN were dissolved in 840 µL of water. 150.9 mg of Pb(NO3)2 was separately dissolved in 1.5 mL of water. The Pb(NO3)2 solution was placed in an ice bath while stirring, and the KSCN solution was injected. The solution was stirred for 30 minutes, and a colorless powder of Pb(SCN)2 with S12CN:S13CN=5:95 was formed. The Pb(SCN)2 was filtered by suction filtration and dried overnight. 20.3 mg of the isotopically mixed Pb(SCN)2 and 20.3 mg of (CH3NH3)I were placed in a vial, and 1.6 mL of THF was added. The solution was sonicated for 10 minutes to dissolve both the solids to yield a yellow solution. The solution was filtered with an inorganic membrane filter (0.1µm) and used for the spin-coating.

8  

ACS Paragon Plus Environment

Page 8 of 33

Page 9 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

The precursor solution and the SiO2/CaF2 substrate were transferred to a nitrogen glove box. 200 µL of the solution was loaded on the substrate and spin-coated using a home-built spincoater at 3000 rpm for 30 seconds. A scarlet film of the Pb-I-SCN layered perovskite was formed on the substrate. As noted previously, the Pb-I-SCN film decomposes in ambient humidity.35 To avoid exposure to atmospheric humidity, the substrate was spin-coated in and kept in a nitrogen glove box. The sample was sealed in the sample cell (sample substrate, CaF2 window and 1 mmthick O-ring, Figure 1B). The powder x-ray diffraction (PXRD) pattern, UV-Vis absorption spectrum, and photoluminescence spectrum of the Pb-I-SCN spin-coated films were reported in a previous publication.35 These sample characterizations are presented and discussed in the Supporting Information. The final measurements of 2D IR lifetimes and spectral diffusion were conducted on samples that were prepared both for the PXRD measurements and the IR experiments in the identical manner. Due to the sensitivity of the sample to the humidity, we could not characterize the sample thickness using a standard technique such as atomic force microscopy (AFM). Instead, we determined the thickness of the films using CN stretch absorption in transmission FT-IR spectra, which shows the thickness of the films to be in the range of 50 nm to 100 nm. See Supporting Information for details. B. Ultrafast Infrared Laser System and the Near-Brewster’s Angle Reflection Geometry The laser system and the implementation of near-Brewster’s angle reflection pump-probe geometry were described in detail in a previous publication.39 An infrared pulse (30 µJ, ~160 fs fwhm, ~95 cm-1 fwhm, centered at 2080 cm-1) was split into a pump pulse (92%) and a probe pulse (8%). The pump pulse was sent to a germanium acousto-optic modulator (Ge-AOM) pulse shaping system, both to control the temporal profile of the pump pulses and to filter some of the

9  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

frequency components as discussed below. The pump pulses were focused into the sample with a f =15 cm lens, and the probe pulse was focused with a f =10 cm lens. The probe pulse arrived after the pump pulses with a controlled delay time. Excessive energy of the pump pulses degraded the samples. Therefore, pump intensity at focus was thus attenuated to ~30 mJ/cm2 to avoid sample damage. The sample (Figure 1B) was placed so that the incident angle of the probe pulse was 53°, which is close to Brewster’s angle for CaF2, 54.5°. The Pb-I-SCN sample layer was placed at the focus of the pump/probe pulses. Then the reflected probe beam from the surface containing the Pb-I-SCN film (the third surface from the top as shown in Figure 1B) was selected and collimated with another f = 10 cm lens, sent to a spectrograph, and detected by an HgCdTe array detector. Note that there are four reflections of the probe pulse caused by the four surfaces. Though these reflections are tightly spaced right after the sample, the collimating lens steers these reflections differently. Consequently, all the reflections were separated by >2 cm in front of the monochromator. Therefore, the influence of the other reflections on the detected signal was eliminated. C. Spectral Filtering of Pump Pulse The AOM pulse shaper can remove some of the frequency components of the incident pump pulses. The details are outlined in previous publications.45, 47 When we addressed the coupling between 12CN and 13CN vibrations in adjacent anions, we employed pump pulses with the full spectrum. We will call this pumping scheme “full pump” in the following discussions. On the other hand, the coupling or heating induced by pumping the S13CN anions, the 95% majority component, hinder the extraction of the dynamical information stored in the observed pump-probe or 2D spectra. To circumvent the problem, the pump spectrum was often modified

10  

ACS Paragon Plus Environment

Page 10 of 33

Page 11 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

by removing the spectral region where the 13CN vibration is located (2015 to 2055 cm-1).45, 48 The spectral shaping with a Hann window was employed to avoid the effects of hard edges of the pump spectrum. This pumping scheme will be called the “(spectrally) filtered pump” (see Figure 2A). Note that by filtering the pump spectral profile, the pump pulse is slightly broadened in time. Careful attention must be paid to avoid potential contributions from non-resonant signals or signals arising from unwanted time-ordered pathways such as perturbed free-induction decays.49 Based on the pump-probe SHG cross-correlation we acquired with the spectrally filtered pump pulse,45 we determined that the pump-probe or 2D IR signals acquired after Tw = 2 ps are free from these potential artifacts. Therefore, the pump-probe signals (Figure 2B) and 2D IR band shapes (Figure 4B) acquired with the spectrally filtered pump pulse were analyzed only after Tw = 2 ps. III. Infrared Spectroscopy on the Layered Perovskite: A. Infrared Absorption Spectroscopy The black lines in Figure 2A show a typical infrared absorption spectrum acquired in the standard transmission geometry. Our main focus is the 2080 cm-1 band arising from the 12CN stretch of the minor S12CN (5%) anion component in the isotopically diluted layered perovskite. As seen in the inset, this 12CN band is a single peak. In contrast, the major 13CN stretch from S13CN (95%) clearly consists of two bands, i.e. the major band at ~2037 cm-1 and a large shoulder on the red side at ~2030 cm-1. Accurate determination of the peak positions was difficult due to the small separation of the two bands.

11  

ACS Paragon Plus Environment

Journal of the American Chemical Society

1mOD

absorbance

13CN

2070

12CN

2080

2090

10mOD 12CN

2000

B

2050

2100

frequency (cm-1)

1.00

pump intensity (a.u.)

A

pump-probe signal (a.u.)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

12CN

0.75 0.50 0.25

11.6 ps

0.00 -20

0

20

t (ps)

40

60

Figure. 2. A. Black solid lines – FT-IR absorption spectrum of the isotopically doped Pb-I-SCN film acquired in transmission geometry. While the large band around 2030-2040 cm-1 arises from major S13CN‒ component, the minor feature at ~2080 cm-1 originates from minor S12CN‒ anions. Inset – Expanded view of the 12CN absorption band. While the 13CN vibration clearly exhibits a shoulder at ~2030 cm-1, the shoulder is absent in the 12 CN spectrum. This is one of the signatures of the vibrational coupling between adjacent SCN‒ anions. (Red solid line) The spectrum of the “filtered” pump pulse in this region. The pump spectrum was modified by the AOM to avoid excitation of the 13CN mode. In contrast, in the full pump spectrum experiments, the 13CN mode is also pumped as indicated by the dotted magenta line. B. Infrared pump-probe signal acquired at the probe frequency of 2080 cm-1 with the pump spectrum filtered so that the 13CN vibration is not excited. The relatively long lifetime, 11.6 ps, provided more than a 20 ps time window over which the structural evolution was monitored with 2D IR spectroscopy. The pump-probe signal decays to zero, demonstrating that there is little sample heating by the laser.

The two bands in the 13CN stretch spectrum can be explained by the vibrational coupling between two adjacent S13CN anions.50 Coupling between the CN vibrations of the closely spaced S13CN anions leads to a splitting of the absorption band, which is equivalent to the well-known Davydov splitting in the electronic spectroscopy of crystals with two or more molecules in the unit cell. The splitting of the 13CN vibrational band can be regarded as a vibrational version of Davydov splitting. The splitting is not observed in the S12CN band, as S12CN is diluted in the structure, and it is not statistically likely that there are two adjacent S12CN anions. When an 12  

ACS Paragon Plus Environment

Page 12 of 33

Page 13 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

S12CN is adjacent to a S13CN, the difference in the 12CN and 13CN vibrational frequency is large compared to the coupling. The coupling will produce a very small frequency shift of the 12CN band, but no splitting.51 Isotopic dilution has been frequently used to eliminate vibrational excitation transfer in the study of dynamics of water (HOD in H2O or D2O).27, 31 There are two possible vibrational coupling mechanisms, mechanical coupling and transition dipole-transition dipole coupling.50, 52 Mechanical coupling is a result of the anharmonicity of the vibrational potential. The vibration of one mode must be coupled through chemical bonds to another mode of a molecule. In this case, mechanical coupling can only take place between two adjacent SCN anions that are chemically connected through the Pb atom. However, the mechanical coupling is expected to be very weak as the large mass of Pb effectively isolates the CN vibrations of the two anions. An example of a massive atom decoupling vibrations is the fact that the CN stretch of benzonitrile has a lifetime of 7 ps while the CN of cyanoselenobenzene has a lifetime of 300 ps.44, 53 The heavy selenium effectively decouples the CN stretch from the modes of the benzene, blocking effective pathways for vibrational relaxation. Transition dipole-transition dipole coupling is through space.54 The two anions do not have to be on the same molecule. As seen in Figure 1A, there are two adjacent SCN‒ anions in different inorganic layers that are not chemically bound but can be coupled because of their proximity, ~4 Å apart (the distance was measured as the distance between the centers of the two adjacent CN bonds). The transition dipole-transition dipole coupling can be calculated based on the reported crystal structure and the transition dipole moment of the CN stretching mode of the SCN‒. Sodium thiocyanate (NaSCN) was dissolved in dimethyl sulfoxide (DMSO), and an infrared

13  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 33

absorption spectrum was acquired to determine the transition dipole moment magnitude of the CN stretch (|µT | = 9.1 × 10-31 C·m). The transition dipole-transition dipole coupling constant, J, is given by

J





      2  3 1  2  3( 1  r )( 2  r ) r , 4 0 r 1

(1)

   where r is a vector connecting the centers of the two CN bonds, and 1 and 2 are the

transition dipole moment vectors for the two CN vibrations. We take the transition dipole to be along the CN bond. Based on the crystal structure in Figure 1A, the coupling constant J was estimated to be 4.4 cm-1. Because the vibrational frequencies for the two oscillators are identical, the splitting is 2J.51 The calculated splitting is thus 8.8 cm-1, which is quite close to the observed splitting of ~7 cm-1. Note that the transition dipole moment in the perovskite lattice could be weaker than that of SCN‒ in solution due to the covalent nature of Pb-SCN bonding, which may account for the slight overestimation of the splitting compared with the experimental value. For example, ethylSCN has a smaller transition dipole than SCN‒. Nonetheless, the calculation supports the transition dipole-transition dipole coupling mechanism between SCN‒ ligands. Note that the identical coupling calculation was applied to the second closest SCN anion pair, which are ~6 Å apart. The calculation yielded the splitting 2J of 4 cm-1. This smaller splitting is obscured by the intrinsic absorption bandwidth (8 cm-1 fwhm) and was not manifested in either linear or nonlinear spectroscopy experiments. We observed that the ratio of the absorption amplitudes between the ~2037 cm-1 and ~2030 cm-1 bands from the S13CN anions varies from sample to sample. This variance could be explained by different orientations of the crystalline domains in different films. As is the case with most other 2D perovskites, powder X-ray diffraction of the films show preferred

14  

ACS Paragon Plus Environment

Page 15 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

orientation, with the inorganic sheets lying parallel to the substrate (with the sheets stacking along the crystallographic a axis)35. However, the crystallites are randomly ordered along the b and c axes. Note that the eigenstates, the sum and difference of the two coupled vibrational modes, have transition dipole moments that are orthogonal in the crystal. For the measurement of the absorption spectrum, the incident beam was sent almost normal to the surface and therefore the polarization was in the plane of the surface. Depending on the crystal orientations within the plane of the inorganic sheets, the field-dipole interactions for the two modes can be different, which leads to the varying ratios. B. Pump-probe Spectroscopy: Vibrational Lifetime and Heating Effects Figure 2B shows the infrared pump-probe signal acquired at the optical frequency of 2080 cm-1 (12CN stretch) with the spectrally filtered pump pulse (see section IIC) so that the 13

CN stretch is not pumped. The polarization for the pump and probe pulses are parallel. As the

orientational motion of the SCN anions are strongly restricted due to the rigid Pb-SCN bond, the observed pump-probe signal decay is solely attributed to vibrational relaxation.55 There are two important features observed in Figure 2B. First, the decay occurs with the vibrational lifetime of 11.6 ps. This relatively long lifetime of the 12CN stretching mode allows the tracking of the evolution of the two-dimensional band shape in 2D IR spectroscopy. Another important point is that the pump-probe signal decays to zero after long enough time, indicating the absence of the significant heating of the sample. This is another advantage of spectrally filtering the pump pulse so that the majority of the anions (S13CN) is not pumped. When the pump-probe signal was acquired with a pump pulse with the full spectral profile, a strong heating signal was observed, which complicated the interpretation of data, particularly that of 2D IR spectroscopy.31

15  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

C. Two-dimensional Infrared Spectroscopy: Dispersive Line Shape Figure 4A in Section V shows two-dimensional infrared (2D IR) spectra acquired with the spectrally filtered pump pulse. Along the ωm axis, dispersive line shapes contribute to the observed 2D spectra, in addition to absorptive line shapes. The phases of the absorptive and dispersive line shapes are exactly in agreement with the predictions in a previous report.39 The mechanism that gives rise to the dispersive line shape contribution was discussed there in detail as well. Briefly, the signal field emitted inside the film is phase-shifted from the signal from the very top of the film due to the additional optical path, which results in the emergence of the dispersive lineshape.39 While the observed 2D band is not purely absorptive, it was demonstrated previously that the center line slope (CLS) of the band shape is preserved regardless of the contribution from the dispersive lineshape.39 The CLS(Tw) is the normalized frequency-frequency correlation function (FFCF), which reflects the structural dynamics of the medium that give rise to spectral diffusion.33, 56 The FFCF decays in time as structural fluctuations occur and randomize the vibrational frequency within the inhomogeneously broadened absorption spectrum; therefore, FFCF is a good measure for the structural dynamics occurring in the film. In the following two sections, 2D IR spectra acquired with the pump pulses with the full (Section IV) and filtered (Section V) spectral profiles are presented to address vibrational coupling and structural dynamics of the perovskite structures, respectively. IV. Vibrational Coupling Revealed by 2D IR Spectroscopy A. Off-Diagonal Cross-Peak

16  

ACS Paragon Plus Environment

Page 16 of 33

Page 17 of 33

A

i

ii

ωm (cm-1)

2080 2060 2040 2040

2080

2060

ωτ (cm-1)

B

ii

ωm (cm-1)

i

0.15

ωτ (cm-1) ~40 cm-1

FFT mag.

C

0.20

CLS

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

0

50

100

freq. (cm-1)

0.10 0

1

2

3 Tw (ps)

4

5

Figure 3. A. 2D IR spectrum of isotopically doped Pb-I-SCN perovskite acquired with the full pump spectrum, which pumps both 12CN and 13CN vibrations. The spectrum was acquired at Tw = 1 ps. B. Expanded views of region (i) and (ii) in the full spectrum shown in panel A. Region (i), which corresponds to the off-diagonal region between 12CN and 13CN vibrations, shows that the positive dispersive signal from the 12CN vibration is eaten out by a negative cross peak between the two vibrations (red arrow). Region (ii) shows the diagonal peak for the 12CN vibration, which can be analyzed with the CLS method to further reveal the vibrational coupling between 12CN and 13CN vibrations. C. The temporal evolution of the CLS for the 12CN diagonal peak shown in B-(ii). The black dots show the raw data, which exhibit an oscillatory feature. The solid line is a cubic spline to the observed data with 10 fs increments. Inset – Fast Fourier Transform magnitude of the splined data. A peak is observed at ~40 cm-1, which corresponds well to the splitting between 12CN and 13CN vibrations.

Figure 3A shows the 2D IR spectrum acquired with the full pump spectrum at Tw = 1 ps. The diagonal peak observed in the 2030-2050 cm-1 region of Figure 3A reveals major and minor features arising from two absorption peaks discussed in Section IIIA, namely the ~2030 cm-1 peak and the ~2037 cm-1 peak. In the 2D IR spectrum, the coupling of the two vibrations produces off-diagonal cross-peaks between the two vibrations. However, the amplitude of the 2030 cm-1 peak is so small compared with the 2037 cm-1 peak that the observation of the crosspeaks is difficult. The contribution of the dispersive line shape further complicates the

17  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

identification of the cross peaks. Therefore, two important subparts of Figure 3A are expanded in Figure 3B providing a path for analysis of the vibrational coupling. To address the coupling between two adjacent vibrations, it is easier in this case to observe the cross-peaks between 12CN and 13CN vibrations, which are well separated in frequency. Note that while the adjacent 12CN and 13CN vibrations do not induce a splitting because of the large separation in the vibrational frequencies,51 they are still coupled through the transition dipole-transition dipole interaction, and cross-peaks should be observed in the 2D spectra. The cross-peak is most clearly visible in panel (i) in Figure 3A and B. The positive (red) feature arises from the dispersive line shape of the major 13CN vibration. Clearly, the positive feature is reduced right at ωm = 2080 cm-1, where the 12CN vibration is located (see arrow). The negative cross peak eats a hole in the positive going peak. This feature shows the existence of the negative-going cross-peak between the 12CN vibration (ωτ = 2037 cm-1) and the 13CN vibration (ωm = 2080 cm-1). The observation of the off-diagonal cross-peak verifies that the two adjacent CN vibrations are coupled. However, because of the complexity of the spectrum, it is easier to see the effect of coupling in the time domain as discussed below. B. CLS Oscillations As confirmation of the coupling of the adjacent CN vibrations, we investigated the waiting time (Tw) dependence of the center line slope (CLS) of the diagonal peak arising from 12

CN vibration. The peak corresponds to panel (ii) in Figure 3A and B. The 2D IR spectra were

again acquired with full spectrum pumping. Figure 3C shows the Tw-dependence of the CLS. The CLS clearly oscillates in time (compare to Figure 4B). Figure 3C inset shows the Fourier transform magnitude spectrum of the

18  

ACS Paragon Plus Environment

Page 18 of 33

Page 19 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

oscillation. The oscillation frequency of ~40 cm-1 can be readily identified, which corresponds very well to the separation between 12CN and 13CN vibrations. This oscillation in CLS is another signature of the vibrational coupling between the two vibrational modes. The mechanism of the CLS oscillation was discussed in detail in a previous publication.45 Briefly, two interactions from the pump pulses each excite both the 12CN and 13CN vibrations, forming a coherence between the two vibrations during the waiting (Tw) period. This pathway introduces a non-rephasing line shape with amplitude and phase oscillating with time, leading to the oscillation in CLS. Overall, the cross-peak and the oscillations in the CLS observed with 2D IR spectroscopy verify that the adjacent CN vibrations are coupled, and the coupling leads to the splitting of the 13

CN vibration absorption band. The two peaks originating from the splitting were indeed

observed in the absorption spectrum (Figure 2A), and are well explained by transition dipoletransition dipole coupling (Section IIIA). When a pure S12CN sample was prepared, the absorption band was also composed of two peaks with the same splitting. Though the oscillation is useful to identify the vibrational coupling, it hinders the accurate assessment of the structural dynamics through the CLS decay. The filtered pumping scheme eliminates the oscillations and yields the 2D band shape evolution caused purely by structural dynamics, as discussed in Section V. V. Structural Fluctuation of the Layered Perovskite To address the structural dynamics occurring in the Pb-I-SCN layered perovskite without interference from vibrational coupling, the frequency filtered pumping scheme was employed to acquire 2D IR spectra for the 12CN vibration. Figure 4A shows two 2D IR spectra acquired at Tw = 2 ps and Tw = 20 ps. A subtle change in the 2D band shape can be observed between the two

19  

ACS Paragon Plus Environment

Journal of the American Chemical Society

B 2090

0.175 0.150

Tw = 2 ps

CLS (Tw)

ωm (cm-1)

A

2080

0.125

0

2090

2080

Tw = 20ps

C

2080

2070 2070

ωτ

5

10

15

20

25

Tw (ps)

2090

potential energy

2070

4.1±0.3 ps

0.100

2070

ωm (cm-1)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 33

2080

2090 -1 (cm )

~4ps

kT potential min.

displacement

12

Figure 4. A. 2D IR spectra of the CN vibration acquired with the spectrally filtered pump pulse so that the 13CN vibration is not pumped. The changes in the slope of the CLS (white dotted lines) with time give the normalized FFCF. B. CLS decay acquired from a series of time-dependent 2D IR spectra. The spectra from three different samples were averaged. The decay time constant, 4.1 ± 0.3 ps, arises from the structural fluctuations of the perovskite structure. C. A schematic of the potential energy surface. The red curve is the potential energy of the periodic lattice. Oscilatory motions about the minimum are the lattice phonons. The observation of the inhomogeneous broadened absorption spectrum and the spectral diffusion shows that there is rough energy landscape with many small potential minima (black curve). Each minimum is a distinct local structure. A particular minimum will be associated with a frequency in the inhomogeneously broadened absorption spectrum. Spectral diffusion is associated with transitions (colored arrows) from one minimum to another.

spectra. The changes in the 2D spectra shapes with Tw are quantified with the CLS method. The dashed white lines in Figure 4A are the CLS. As Tw increases the slope becomes smaller, approaching zero at long times. 2D IR spectra were acquired for the various waiting times between 2 ps and 25 ps, and the CLS values were calculated around the main negative-going feature. The series of 2D IR spectra were acquired from three independent samples, and they yielded very similar CLS decays. The CLS decays were weighted based on the number of scans and averaged to yield an averaged CLS decay with reasonable signal-to-noise ratio as shown in Figure 4B. The CLS

20  

ACS Paragon Plus Environment

Page 21 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

decay in Figure 4B was fit well by a single exponential plus offset, and yielded the time constant of 4.1±0.3 ps with a 0.09 offset. Note that the total decay amplitude is quite small (~0.1). Again, the CLS decay is the normalized FFCF decay, independent of the contribution from the dispersive line shape. The experiments shown in Figure 4B were taken on samples with 5% 12C and 95% 13C SCN. To confirm that coupling between the low concentration of pairs of 12C SCNs, the experiments were repeated with samples with 1% 12C and 99% 13C SCN. The results were identical. When the CLS decay is extrapolated to Tw = 0 ps based on the fit, it yields the value of ~0.2. The maximum possible value is 1. The difference between the observed Tw = 0 value and 1 is a measure of the homogeneous linewidth. This small value of the CLS at Tw = 0 indicates that the 12CN absorption band is substantially homogeneously broadened. Analyzing the data quantitatively, the homogenous linewidth was determined to be 6 cm-1 out of the total 8 cm-1 full width at half maximum (fwhm) of the absorption band. The total linewidth is the convolution of the homogeneous component with the inhomogeneously broadened component. The inhomogeneous width reflects structural disorder in the crystal. Some of this disorder is not static but rather dynamic, unlike disorder caused by, e.g., crystal lattice strains and defects. Such dynamical disorder is observed as the 4.1 ps spectral diffusion, which originates from fast sampling of different structures. Out of the 8 cm-1 fwhm linewidth, the 3.5 cm-1 fwhm inhomogeneous linewidth component is sampled by structural fluctuations. Weak coupling between differences in the lattice structure and the CN frequency will produce a small inhomogeneous component even if the disorder is significant. It is important to recognize that spectral diffusion is not caused by lattice phonons oscillatory motions on the periodic lattice potential (red curve in Figure 4C). Rather, the observed spectral diffusion is caused by

21  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

transitions from one local lattice configuration to another via transitions from one small potential energy minimum to another on the rough energy landscape shown schematically as the black curve in Figure 4C. Different structures give rise to different CN vibrational frequencies, e.g., due to slightly different distance between SCN– anions and CH3NH3+ cations. Thermally driven transitions from one local minimum to another, as indicated in Figure 4C, result in frequency changes, which were detected as the spectral diffusion with the time constant of 4.1 ps. The observation here of the spectral diffusion of the CN vibrational mode of SCN‒ anion demonstrates that SCN‒ anion is part of a dynamical lattice structure in the Pb-I-SCN layered perovskite. The spectral diffusion decay shows that the layered perovskite structure is constantly undergoing fast structural changes with a time constant of 4.1 ± 0.3 ps. The dynamical disorder in the 2D perovskite may play a role in stabilizing excitons soon after they are generated before recombination can take place. As shown by the small offset in the CLS (0.09), a small fraction (2.5 cm-1 fwhm) of the vibrational absorption band (8 cm-1 fwhm band) is not sampled by the structural fluctuations discussed above and appears as static inhomogeneity over the observable time window. As the observable time range is limited by the vibrational lifetime of the SCN (~11 ps), we cannot exclude the possibility that much slower (>100 ps) local structural fluctuations sample this inhomogeneity, leading to complete spectral diffusion. However, considering that the film is polycrystalline and contains various crystals with different sizes and strains, it is possible that the small residual inhomogeneity arises from the static inhomogeneity due to strains, defects and grain boundaries. VI. Concluding Remarks

22  

ACS Paragon Plus Environment

Page 22 of 33

Page 23 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Two-dimensional infrared (2D IR) spectroscopy was applied to a thin film of the Pb-ISCN layered perovskite using near-Brewster’s angle reflection pump-probe geometry. To acquire the correct dynamical information, the SCN‒ anions in the layered perovskite were isotopically diluted, which renders the 12CN absorption of interest as small as ~2 mOD. In spite of the small optical density, the 2D IR spectra acquired in near-Brewster’s angle reflection pump-probe geometry yielded the good quality data shown in Figure 4. The data reveal the existence of fast structural fluctuations (~4 ps) of the layered perovskite. The nature of the rapid structural changes is schematically illustrated in Figure 4C. The 2D IR observation of spectral diffusion and inhomogeneous broadening of the CN stretch absorption spectrum demonstrates that the lattice periodic potential has associated with it a rough energy landscape (Figure 4C). Spectral diffusion reports on the time dependent sampling of local structural minima of the landscape. The structural evolution may be important for the stabilization of optically generated charge carriers or excitons prior to recombination. The results also showed that the CN vibrations on adjacent SCN‒ anions are coupled through transition dipole-transition dipole interactions. Although the inorganic layers are structurally and electronically isolated in typical layered perovskites, here, the layers are directly linked through SCN‒ vibrations. 2D spectroscopy has been applied to other aspects of perovskites. For example, Thouin and co-workers applied 2D visible spectroscopy to another type of 2D perovskite to reveal high biexciton binding energy.57 Applying 2D IR spectroscopy to perovskites requires a vibrational probe. Bakulin and co-workers obtained transmission 2D IR signals from the N-H bending mode of the organic cations in the 3D perovskite (CH3NH3)PbI3 and, based on the polarizationdependent peak volumes, they discussed the rotational dynamics of the organic cations.12, 58 A similar method was recently applied to the CN stretching mode of a cation in formamidinium

23  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

lead iodide perovskite.59 These experiments addressed orientational dynamics with 2D IR but not spectral diffusion. These same vibrational modes could potentially be employed to track the evolution of the 2D line shapes (spectral diffusion) using the reflection method employed here. There are also a number of reports that SCN anions can be doped to replace a fraction of iodides in (CH3NH3)PbI3.60-62 Such doped perovskites might be suitable for studying common 3D systems.

Acknowledgements This material is based upon work and instrumentation supported by the Air Force Office of Scientific Research under AFOSR Award No. FA9550-16-1-0104. Support for JPB was provided by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy through Grant #DE-FG03-84ER13251. Work by HIK, KPL, and DU was supported by the NSF CAREER award DMR-1351538.

24  

ACS Paragon Plus Environment

Page 24 of 33

Page 25 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

References 1. Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T.,Organometal Halide Perovskites as Visible-Light Sensitizers for Photovoltaic Cells. J. Am. Chem. Soc. 2009, 131, 6050-6051. 2. Im, J.-H.; Lee, C.-R.; Lee, J.-W.; Park, S.-W.; Park, N.-G.,6.5% Efficient Perovskite Quantum-Dot-Sensitized Solar Cell. Nanoscale 2011, 3, 4088-4093. 3. Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J.,Efficient Hybrid Solar Cells Based on Meso-Superstructured Organometal Halide Perovskites. Science 2012, 338, 643647. 4. Bryant, D.; Aristidou, N.; Pont, S.; Sanchez-Molina, I.; Chotchunangatchaval, T.; Wheeler, S.; Durrant, J. R.; Haque, S. A.,Light and Oxygen Induced Degradation Limits the Operational Stability of Methylammonium Lead Triiodide Perovskite Solar Cells. Eng. & Env. Sci. 2016, 9, 1655-1660. 5. Chun-Ren Ke, J.; Walton, A. S.; Lewis, D. J.; Tedstone, A.; O'Brien, P.; Thomas, A. G.; Flavell, W. R.,In Situ Investigation of Degradation at Organometal Halide Perovskite Surfaces by X-Ray Photoelectron Spectroscopy at Realistic Water Vapour Pressure. Chem. Commun. 2017, 53, 5231-5234. 6. Yang, W. S.; Park, B.-W.; Jung, E. H.; Jeon, N. J.; Kim, Y. C.; Lee, D. U.; Shin, S. S.; Seo, J.; Kim, E. K.; Noh, J. H.; Seok, S. I.,Iodide Management in Formamidinium-Lead-Halide– Based Perovskite Layers for Efficient Solar Cells. Science 2017, 356, 1376-1379. 7. Arora, N.; Dar, M. I.; Hinderhofer, A.; Pellet, N.; Schreiber, F.; Zakeeruddin, S. M.; Grätzel, M.,Perovskite Solar Cells with Cuscn Hole Extraction Layers Yield Stabilized Efficiencies Greater Than 20%. Science 2017.

25  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

8. Kitazawa, N.,Excitons in Two-Dimensional Layered Perovskite Compounds: (C6H5C2H4NH3)2Pb (Br, I)4 and (C6H5C2H4NH3)2Pb(Cl, Br)4. Mater. Sci. Eng. B 1997, 49, 233238. 9. Dohner, E. R.; Jaffe, A.; Bradshaw, L. R.; Karunadasa, H. I.,Intrinsic White-Light Emission from Layered Hybrid Perovskites. J. Am. Chem. Soc. 2014, 136, 13154-13157. 10. Poglitsch, A.; Weber, D.,Dynamic Disorder in Methylammoniumtrihalogenoplumbates (Ii) Observed by Millimeter‐Wave Spectroscopy. J. Chem. Phys. 1987, 87, 6373-6378. 11. Yaffe, O.; Guo, Y.; Tan, L. Z.; Egger, D. A.; Hull, T.; Stoumpos, C. C.; Zheng, F.; Heinz, T. F.; Kronik, L.; Kanatzidis, M. G.,Local Polar Fluctuations in Lead Halide Perovskite Crystals. Phys. Rev. Lett. 2017, 118, 136001. 12. Bakulin, A. A.; Selig, O.; Bakker, H. J.; Rezus, Y. L.; Müller, C.; Glaser, T.; Lovrincic, R.; Sun, Z.; Chen, Z.; Walsh, A.,Real-Time Observation of Organic Cation Reorientation in Methylammonium Lead Iodide Perovskites. J. Phys. Chem. Lett 2015, 6, 3663-3669. 13. Wu, X.; Tan, L. Z.; Shen, X.; Hu, T.; Miyata, K.; Trinh, M. T.; Li, R.; Coffee, R.; Liu, S.; Egger, D. A.,Light-Induced Picosecond Rotational Disordering of the Inorganic Sublattice in Hybrid Perovskites. Sci. Adv. 2017, 3, e1602388. 14. Stranks, S. D.; Eperon, G. E.; Grancini, G.; Menelaou, C.; Alcocer, M. J. P.; Leijtens, T.; Herz, L. M.; Petrozza, A.; Snaith, H. J.,Electron-Hole Diffusion Lengths Exceeding 1 Micrometer in an Organometal Trihalide Perovskite Absorber. Science 2013, 342, 341-344. 15. D’Innocenzo, V.; Grancini, G.; Alcocer, M. J. P.; Kandada, A. R. S.; Stranks, S. D.; Lee, M. M.; Lanzani, G.; Snaith, H. J.; Petrozza, A.,Excitons Versus Free Charges in Organo-Lead TriHalide Perovskites. Nat. Comm. 2014, 5, 3586.

26  

ACS Paragon Plus Environment

Page 26 of 33

Page 27 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

16. Collavini, S.; Völker, S. F.; Delgado, J. L.,Understanding the Outstanding Power Conversion Efficiency of Perovskite-Based Solar Cells. Angew. Chem. Int. Ed. 2015, 54, 97579759. 17. Jha, A.; Duan, H.-G.; Tiwari, V.; Nayak, P. K.; Snaith, H. J.; Thorwart, M.; Miller, R. J. D.,Direct Observation of Ultrafast Exciton Dissociation in Lead Iodide Perovskite by 2D Electronic Spectroscopy. ACS Phot. 2017. 18. Ziffer, M. E.; Mohammed, J. C.; Ginger, D. S.,Electroabsorption Spectroscopy Measurements of the Exciton Binding Energy, Electron–Hole Reduced Effective Mass, and Band Gap in the Perovskite Ch3nh3pbi3. ACS Phot. 2016, 3, 1060-1068. 19. Zhu, H.; Miyata, K.; Fu, Y.; Wang, J.; Joshi, P. P.; Niesner, D.; Williams, K. W.; Jin, S.; Zhu, X.-Y.,Screening in Crystalline Liquids Protects Energetic Carriers in Hybrid Perovskites. Science 2016, 353, 1409-1413. 20. Zheng, F.; Tan, L. Z.; Liu, S.; Rappe, A. M.,Rashba Spin–Orbit Coupling Enhanced Carrier Lifetime in CH3NH3PbI3. Nano Lett. 2015, 15, 7794-7800. 21. Muljarov, E.; Tikhodeev, S.; Gippius, N.; Ishihara, T.,Excitons in Self-Organized Semiconductor/Insulator Superlattices: Pbi-Based Perovskite Compounds. Phys. Rev. B 1995, 51, 14370. 22. Smith, M.; Karunadasa, H.,White-Light Emission from Layered Halide Perovskites. Acc. Chem. Res. 2018. 23. Miyata, K.; Meggiolaro, D.; Trinh, M. T.; Joshi, P. P.; Mosconi, E.; Jones, S. C.; De Angelis, F.; Zhu, X.-Y.,Large Polarons in Lead Halide Perovskites. Sci. Adv. 2017, 3, e1701217. 24. Hamm, P.; Zanni, M. T., Concepts and Methods of 2D Infrared Spectroscopy. Cambridge University Press: New York, 2011.

27  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

25. Woutersen, S.; Mu, Y.; Stock, G.; Hamm, P.,Subpicosecond Conformational Dynamics of Small Peptides Probed by Two-Dimensional Vibrational Spectroscopy. Proc. Nat. Ac. Sci. 2001, 98, 11254-11258. 26. Mukherjee, P.; Kass, I.; Arkin, I. T.; Zanni, M. T.,Picosecond Dynamics of a Membrane Protein Revealed by 2D IR. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 3528-3533. 27. Fecko, C.; Eaves, J.; Loparo, J.; Tokmakoff, A.; Geissler, P.,Ultrafast Hydrogen-Bond Dynamics in the Infrared Spectroscopy of Water. Science 2003, 301, 1698-1702. 28. Kim, Y. S.; Hochstrasser, R. M.,Chemical Exchange 2D IR of Hydrogen-Bond Making and Breaking. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 11185-11190. 29. Zheng, J. R.; Kwak, K.; Asbury, J.; Chen, X.; Piletic, I. R.; Fayer, M. D.,Ultrafast Dynamics of Solute-Solvent Complexation Observed at Thermal Equilibrium in Real Time. Science 2005, 309, 1338-1343. 30. King, J. T.; Kubarych, K. J.,Site-Specific Coupling of Hydration Water and Protein Flexibility Studied in Solution with Ultrafast 2D-IR Spectroscopy. J. Am. Chem. Soc. 2012, 134, 18705-18712. 31. Asbury, J. B.; Steinel, T.; Stromberg, C.; Corcelli, S. A.; Lawrence, C. P.; Skinner, J. L.; Fayer, M. D.,Water Dynamics: Vibrational Echo Correlation Spectroscopy and Comparison to Molecular Dynamics Simulations. J. Phys. Chem. A 2004, 108, 1107-1119. 32. Park, S.; Kwak, K.; Fayer, M. D.,Ultrafast 2D-IR Vibrational Echo Spectroscopy: A Probe of Molecular Dynamics. Laser Phys. Lett. 2007, 4, 704-718. 33. Kwak, K.; Park, S.; Finkelstein, I. J.; Fayer, M. D.,Frequency-Frequency Correlation Functions and Apodization in Two-Dimensional Infrared Vibrational Echo Spectroscopy: A New Approach. J. Chem. Phys. 2007, 127, 124503.

28  

ACS Paragon Plus Environment

Page 28 of 33

Page 29 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

34. Daub, M.; Hillebrecht, H.,Synthesis, Single‐Crystal Structure and Characterization of (Ch3nh3) 2pb (Scn) 2i2. Angew. Chem. Int. Ed. 2015, 54, 11016-11017. 35. Umeyama, D.; Lin, Y.; Karunadasa, H. I.,Red-to-Black Piezochromism in a Compressible Pb–I–Scn Layered Perovskite. Chem. Mater. 2016, 28, 3241-3244. 36. Xiao, Z.; Meng, W.; Saparov, B.; Duan, H.-S.; Wang, C.; Feng, C.; Liao, W.; Ke, W.; Zhao, D.; Wang, J.,Photovoltaic Properties of Two-Dimensional (CH3NH3)2Pb(SCN)2I2 Perovskite: A Combined Experimental and Density Functional Theory Study. J. Phys. Chem. Lett. 2016, 7, 1213-1218. 37. Jaffe, A.; Lin, Y.; Karunadasa, H. I.,Halide Perovskites under Pressure: Accessing New Properties through Lattice Compression. ACS En. Lett. 2017, 2, 1549-1555. 38. Im, J.-H.; Jang, I.-H.; Pellet, N.; Grätzel, M.; Park, N.-G.,Growth of CH3NH3PbI3 Cuboids with Controlled Size for High-Efficiency Perovskite Solar Cells. Nature Nano 2014, 9, 927-932. 39. Nishida, J.; Yan, C.; Fayer, M. D.,Enhanced Nonlinear Spectroscopy for Monolayers and Thin Films in near-Brewster’s Angle Reflection Pump-Probe Geometry. J. Chem. Phys. 2017, 146, 094201. 40. Shim, S.-H.; Zanni, M. T., How to Turn Your Pump-Probe Instrument into a Multidimensional Spectrometer: 2D IR and Vis Spectroscopies Via Pulse Shaping. PCCP 2009, 11, 748-761. 41. Kumar, S. K. K.; Tamimi, A.; Fayer, M. D.,Comparisons of 2D IR Measured Spectral Diffusion in Rotating Frames Using Pulse Shaping and in the Stationary Frame Using the Standard Method. J. Chem. Phys. 2012, 137, 184201. 42. Yan, C.; Thomaz, J. E.; Wang, Y.-L.; Nishida, J.; Yuan, R.; Breen, J. P.; Fayer, M. D.,Ultrafast to Ultraslow Dynamics of a Langmuir Monolayer at the Air/Water Interface

29  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Observed with Reflection Enhanced 2D IR Spectroscopy. J. Am. Chem. Soc. 2017, 139, 1651816527. 43. Ren, Z.; Brinzer, T.; Dutta, S.; Garrett-Roe, S.,Thiocyanate as a Local Probe of Ultrafast Structure and Dynamics in Imidazolium-Based Ionic Liquids: Water-Induced Heterogeneity and Cation-Induced Ion Pairing. J. Phys. Chem B 2015, 119, 4699-4712. 44. Sokolowsky, K. P.; Fayer, M. D.,Dynamics in the Isotropic Phase of Nematogens Using 2D IR Vibrational Echo Measurements on Natural-Abundance 13CN and Extended Lifetime Probes. J. Phys. Chem B 2013, 117, 15060-15071. 45. Nishida, J.; Tamimi, A.; Fei, H.; Pullen, S.; Ott, S.; Cohen, S. M.; Fayer, M. D.,Structural Dynamics inside a Functionalized Metal–Organic Framework Probed by Ultrafast 2D IR Spectroscopy. Proc. Nat. Ac. Sci. 2014, 111, 18442-18447. 46. Rosenfeld, D. E.; Gengeliczki, Z.; Smith, B. J.; Stack, T. D. P.; Fayer, M. D.,Structural Dynamics of a Catalytic Monolayer Probed by Ultrafast 2D IR Vibrational Echoes. Science 2011, 334, 634-639. 47. Giammanco, C. H.; Kramer, P. L.; Yamada, S. A.; Nishida, J.; Tamimi, A.; Fayer, M. D.,Coupling of Carbon Dioxide Stretch and Bend Vibrations Reveals Thermal Population Dynamics in an Ionic Liquid. J. Phys. Chem B 2016, 120, 549-556. 48. Marroux, H. J.; Orr-Ewing, A. J.,Distinguishing Population and Coherence Transfer Pathways in a Metal Dicarbonyl Complex Using Pulse-Shaped Two-Dimensional Infrared Spectroscopy. J. Phys. Chem B 2016, 120, 4125-4130. 49. Yan, S.; Seidel, M. T.; Tan, H.-S.,Perturbed Free Induction Decay in Ultrafast Mid-IR Pump–Probe Spectroscopy. Chem. Phys. Lett. 2011, 517, 36-40. 50. Herzberg, G., Molecular Spectra and Molecular Structure. Read Books Ltd: 2013; Vol. 1.

30  

ACS Paragon Plus Environment

Page 30 of 33

Page 31 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

51. Wong, D. B.; Giammanco, C. H.; Fenn, E. E.; Fayer, M. D.,Dynamics of Isolated Water Molecules in a Sea of Ions in a Room Temperature Ionic Liquid. J. Phys. Chem. B 2013, 117, 623-635. 52. Golonzka, O.; Khalil, M.; Demirdöven, N.; Tokmakoff, A.,Vibrational Anharmonicities Revealed by Coherent Two-Dimensional Infrared Spectroscopy. Phys. Rev. Lett. 2001, 86, 2154. 53. Bian, H.; Wen, X.; Li, J.; Zheng, J.,Mode-Specific Intermolecular Vibrational Energy Transfer. Ii. Deuterated Water and Potassium Selenocyanate Mixture. J. Chem. Phys. 2010, 133, 034505. 54. Eigner, A. A.; Konold, P. E.; Massari, A. M.,Infrared Spectroscopic Signatures of Phase Segregation in P3ht-Porphyrin Blends. J. Phys. Chem. B 2009, 113, 14549-14554. 55. Tokmakoff, A.,Orientational Correlation Functions and Polarization Selectivity for Nonlinear Spectroscopy of Isotropic Media. I. Third Order. J. Chem. Phys. 1996, 105, 1-12. 56. Kwak, K.; Rosenfeld, D. E.; Fayer, M. D.,Taking Apart the Two-Dimensional Infrared Vibrational Echo Spectra: More Information and Elimination of Distortions. J. Chem. Phys. 2008, 128, 204505. 57. Thouin, F.; Neutzner, S.; Cortecchia, D.; Dragomir, V. A.; Soci, C.; Salim, T.; Lam, Y. M.; Leonelli, R.; Petrozza, A.; Kandada, A. R. S.; Silva, C.,Stable Biexcitons in Two-Dimensional Metal-Halide Perovskites with Strong Dynamic Lattice Disorder. Phys. Rev. Mat. 2018, 2, 034001. 58. Selig, O.; Sadhanala, A.; Muller, C.; Lovrincic, R.; Chen, Z.; Rezus, Y. L.; Frost, J. M.; Jansen, T. L.; Bakulin, A. A.,Organic Cation Rotation and Immobilization in Pure and Mixed Methylammonium Lead-Halide Perovskites. J. Am. Chem. Soc. 2017, 139, 4068-4074.

31  

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

59. Taylor, V. C. A.; Tiwari, D.; Duchi, M.; Donaldson, P. M.; Clark, I. P.; Fermin, D. J.; Oliver, T. A. A.,Investigating the Role of the Organic Cation in Formamidinium Lead Iodide Perovskite Using Ultrafast Spectroscopy. J. Phys. Chem. Lett. 2018, 9, 895-901. 60. Tai, Q.; You, P.; Sang, H.; Liu, Z.; Hu, C.; Chan, H. L.; Yan, F.,Efficient and Stable Perovskite Solar Cells Prepared in Ambient Air Irrespective of the Humidity. Nat. Comm. 2016, 7. 61. Halder, A.; Chulliyil, R.; Subbiah, A. S.; Khan, T.; Chattoraj, S.; Chowdhury, A.; Sarkar, S. K.,Pseudohalide (SCN−)-Doped MAPbI3 Perovskites: A Few Surprises. J. Phys. Chem. Lett 2015, 6, 3483-3489. 62. Chen, Y.; Li, B.; Huang, W.; Gao, D.; Liang, Z.,Efficient and Reproducible CH3NH3PbI3−X(SCN)X Perovskite Based Planar Solar Cells. Chem. Commun. 2015, 51, 1199711999.

 

32  

ACS Paragon Plus Environment

Page 32 of 33

Page 33 of 33 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

TOC Graphic

33  

ACS Paragon Plus Environment